Large dynamic pressure aerodynamic force / heat / structure / control multi-field coupling calculation and control method

Abstract

The invention discloses a large dynamic pressure aerodynamic force / aerodynamic heat / structure / control multi-field coupling calculation and control method. The method comprises the following steps: calculating to obtain aerodynamic force / aerodynamic heat / structure multi-field coupling characteristics of a large dynamic pressure aircraft; aerodynamic model correction is carried out according to multi-field coupling characteristics, and a rigid-elastic coupling dynamic model of the large dynamic pressure aircraft is established; designing a sliding mode surface capable of ensuring the stability of a rigid-elastic coupling mode, and obtaining an equivalent control item based on a rigid-elastic coupling dynamic model; adding a super-spiral sliding mode switching control item to form a super-spiral sliding mode closed-loop control law; and designing an output layer weight adaptive law of the BRF neural network state observer, realizing online estimation of a rigid-elastic coupling mode, and forming a super-spiral sliding mode closed-loop control law based on intelligent online estimation of the rigid-elastic coupling mode for a high dynamic pressure flight environment. According to the method, the aerodynamic / thermal / structural multi-field coupling characteristics of the aircraft in the large dynamic pressure environment can be accurately obtained, and finally the purpose of high-precision and high-robustness control in the large dynamic pressure flight environment is achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of aircraft and relates to a large dynamic pressure aerodynamic / thermal / structural / control multi-field coupling calculation and control method. Background Art

[0002] In order to adapt to long-term flight in a high dynamic pressure environment, high dynamic pressure aircraft generally adopt a high slenderness ratio low drag aerodynamic design and a lightweight body structure design. Such aircraft also face harsh and changeable aerodynamic / thermal environments during flight, and the thermal loads they are subjected to are far more severe than those of traditional aircraft, which brings severe challenges to aircraft design. On the one hand, the lightweight body structure is easily affected by the coupling of aerodynamic / thermal loads (i.e. aerodynamic heat) due to its low stiffness, and the elastic deformation of the structure will change the aircraft flow field and thus affect the flow characteristics, showing a complex aerodynamic effect. The complex interference caused by the mutual coupling of the two on the aircraft body cannot be ignored; on the other hand, the dynamic characteristics and control response of the aircraft are significantly changed under the action of multi-field coupling, especially the influence of the rigid-elastic coupling effect. The modeling and simulation of the spatiotemporal evolution of the dynamic response of the aircraft is extremely complex, and the traditional control method based on rigid body dynamics modeling is no longer applicable, which seriously affects the control performance and thus restricts the flight performance.

[0003] Although domestic and foreign scholars have conducted some research on the aerodynamic / thermal / structural / control multi-field coupling problem, they mainly focus on low dynamic pressure flight environments. The effectiveness of numerical simulation of multi-field coupling characteristics in high dynamic pressure environments has not been fully verified. It is necessary to make adaptive improvements to the simulation methods and processes of multi-field coupling characteristics in traditional low dynamic pressure environments. In addition, the aeroelastic effect is aggravated in high dynamic pressure environments, and the trim attitude and stability characteristics of the aircraft may change significantly. In dynamic modeling, it is necessary to accurately describe the dynamic process of the elastic mode and the impact on the dynamic characteristics. Therefore, in the process of control law design, it is necessary to take targeted solutions to the complex effects of high dynamic pressure multi-field coupling. Summary of the invention

[0004] The present invention aims to solve at least one of the problems existing in the prior art.

[0005] To this end, the present invention provides a large dynamic pressure aerodynamic / thermal / structural / control multi-field coupling calculation and control method. The method is closely centered on engineering practice and aims to accurately obtain the aerodynamic / thermal / structural multi-field coupling characteristics of the aircraft in a large dynamic pressure environment, and complete the rigid-elastic coupling dynamics modeling of the large dynamic pressure aircraft based on the coupling characteristics, and then design a super-helical sliding mode closed-loop control law based on the intelligent online estimation of the rigid-elastic coupling mode that can suppress the complex influence of multi-field coupling, and finally achieve the purpose of high-precision and strong robust control in a large dynamic pressure flight environment.

[0006] The technical solution of the present invention is:

[0007] A large dynamic pressure aerodynamic / thermal / structural / control multi-field coupling calculation and control method, the specific steps are:

[0008] In the first step, the coupled numerical simulation method of "CFD in improved spatial discretization format + CSD with optimized grid based on load distribution" is used to calculate the elastic deformation data of the high dynamic pressure aircraft under the action of aerodynamic force and aerothermal force, and then the multi-field coupling characteristics of aerodynamic force / aerothermal force / structure are obtained;

[0009] The second step is to establish the aerodynamic model and dynamic model of the aircraft, carry out the aerodynamic model correction according to the multi-field coupling characteristics, introduce the elastic state quantity of the aircraft into the aerodynamic model polynomial, and introduce the elastic vibration equation into the dynamic model, and then establish the rigid-elastic coupling dynamic model of the large dynamic pressure aircraft;

[0010] The third step is to design a sliding surface that can ensure the stabilization of the rigid-elastic coupling mode, and obtain the equivalent control term based on the rigid-elastic coupling dynamics model; further add the super-helical sliding mode switching control term to form a super-helical sliding mode closed-loop control law;

[0011] The fourth step is to design the output layer weight adaptation law of the BRF neural network state observer to realize the online estimation of the rigid-elastic coupling mode, and finally form a super-helical sliding mode closed-loop control law based on the intelligent online estimation of the rigid-elastic coupling mode for the high dynamic pressure flight environment.

[0012] Furthermore, the specific steps of the first step are as follows:

[0013] Step 1-1, for a large dynamic pressure aircraft, a fluid domain model grid and a solid domain model grid are simultaneously established, and the multi-field grids, that is, the fluid domain model grid and the solid domain model grid share geometric shape boundaries;

[0014] Step 1-2, numerically simulating the fluid domain model grid to obtain numerical simulation results; in the process of numerically simulating the fluid domain model grid, improving the spatial discrete format of the numerical method of the simulation process;

[0015] Step 1-3, carry out load analysis based on the numerical simulation results obtained in step 1-2 to obtain aerodynamic and aerodynamic thermal data under high dynamic pressure flight conditions; in the process of load analysis, optimize the mesh of the fluid domain model and the mesh of the solid domain model;

[0016] Steps 1-4, conduct aerodynamic thermal-structural heat transfer simulation analysis on the optimized multi-field grid to obtain the wall temperature field results of the solid domain model grid under aerodynamic heating conditions;

[0017] Step 1-5, based on the aerodynamic data obtained in step 1-3 and the temperature field results obtained in step 1-4, simulate and analyze the multi-field coupling characteristics under high dynamic pressure flight conditions to obtain the elastic deformation data of the aircraft; in the simulation and analysis process, the severe area of ​​the aerodynamic force and temperature field uses a direct simulation method to directly map the node distribution data of the aerodynamic force and temperature field on the grid boundary of the solid domain model, and the non-severe area uses a load migration method of conservative flux interpolation to interpolate the aerodynamic force and temperature field of the fluid domain model grid to the grid boundary of the solid domain model;

[0018] Steps 1-6, for the elastic deformation data of the aircraft, use the topology-preserving dynamic mesh deformation technology to obtain the deformed fluid domain model mesh and solid domain model mesh, and iteratively perform elastic coupling calculations until the final deformation converges to obtain the multi-field coupling characteristics under high dynamic pressure flight conditions.

[0019] Furthermore, in step 1-2, the improved spatial discretization format is shown in formula (1):

[0020]

[0021] in: is the numerical flux, d k is the linear weight coefficient, ω k is the nonlinear weight coefficient, k=0,1,2.

[0022] Furthermore, in steps 1-3, the optimization process of the fluid domain model mesh and the solid domain model mesh is: splitting the multi-field mesh area, retaining the multi-field mesh in the harsh area of ​​force / thermal load, and interpolating and sparsely analyzing the number of boundary mesh nodes in the non-harsh area.

[0023] Furthermore, the specific steps of the second step are as follows:

[0024] Step 2-1, based on the elastic deformation data obtained in the first step, use the natural vibration mode Φ i The linear combination of (x), i=1,2,...,∞ expresses the elastic deformation function:

[0025]

[0026] Where: y(x,t) is the elastic deformation function of the body, q i (t) is the generalized coordinate of the i-th elastic mode;

[0027] The rigid-elastic coupling dynamic equations of the high dynamic pressure aircraft are established; where i is the order of the natural vibration mode, x is the x-axis position of the point on the fuselage from the origin of the coordinate system in the body coordinate system, x is a continuous value, and Φ i (x) is the i-th order natural vibration mode at point x;

[0028] Step 2-2, based on the aerodynamic data obtained in the first step, assuming that the aerodynamic forces generated by the fuselage and wings of the aircraft are treated as distributed forces, and the control forces generated by the thrust and control surfaces are treated as concentrated forces, the aerodynamic force F is calculated. y , aerodynamic moment M zy , the concentrated force F generated by the thrust P and thrust moment M zP ;

[0029] Step 2-3, introduce the local angle of attack α K And control rudder deflection δ K , corrected aerodynamic force and aerodynamic moment;

[0030] In step 2-4, based on the small disturbance assumption, the coupling between modes is ignored, and the rigid-elastic coupled dynamic equations of the large dynamic pressure aircraft established in step 2-1 are subjected to Lagrangian treatment to obtain the rigid-elastic coupled dynamic equations of the large dynamic pressure aircraft used for control system design.

[0031] Furthermore, in step 2-1, based on the Lagrange equation, the separation of variables method is used to solve the micro-element motion equation, which is combined with the center of mass translation motion equation and the rotation equation around the center of mass to obtain the rigid-elastic coupled dynamic equation group of the high dynamic pressure aircraft:

[0032]

[0033] Where m is the mass of the aircraft, V is the flight speed, θ is the track angle, and F is the normal force. is the pitch angle, J z is the moment of inertia of the aircraft about the z-axis of the body coordinate system, M z is the pitching moment, q i (t) is the generalized coordinate of the i-th elastic mode, ξ i and ω i are the structural damping ratio and natural frequency of the i-th elastic mode, Q i is the generalized force, M i is the generalized mass, F is the distributed aerodynamic force F y The concentrated force F generated by the thrust P The normal force, M z is the aerodynamic moment M zy With thrust torque M zP The pitching moment can be expressed as:

[0034] F=F y +F P (4)

[0035] M z =M zy +M zP (5)

[0036] The expressions of generalized mass and generalized force are:

[0037]

[0038] Where L is the total length of the aircraft fuselage, m(x) is the distribution of the aircraft mass along the fuselage direction, and f(x, t) is the distribution of the time-varying normal force along the fuselage direction.

[0039] Furthermore, in step 2-2, the concentrated force F generated by the thrust P and thrust moment M zP They are:

[0040]

[0041] Where P is thrust, α is the rigid attack angle, and x p is the x-axis coordinate of the engine thrust point with the leading edge of the aircraft as the coordinate origin, l c is the distance from the engine nozzle section to the center of mass of the fuselage, l c =x p -x CM , x CM is the x-axis coordinate of the center of mass of the aircraft, Φ i (x p ) is x p The i-th order natural vibration mode at point , and:

[0042]

[0043] Aerodynamic force F y for:

[0044]

[0045] F y The aerodynamic torque M generated zy for:

[0046]

[0047] Where ρ is the atmospheric density, S is the reference area of ​​the aircraft, and C y is the lift coefficient, α is the rigid angle of attack, α K is the local angle of attack; δ K To control the rudder deviation, δ z The rudder deflection required by the control signal, is the x-axis coordinate of the point of action of the control force generated by the rudder surface; Δ(·) is the Diracδ function;

[0048] In step 2-3, the local angle of attack α KIt consists of three parts: the rigid angle of attack α, the angle of attack α1 caused by the rotation of the rigid body, and the additional angle of attack α2 caused by elastic deformation. The total local angle of attack is:

[0049]

[0050] In the formula, is the pitch angle rate, Φ i (x) is the i-th order natural vibration mode at point x;

[0051] The control rudder deflection δ K It consists of three parts: the rudder deflection required by the control signal z , the additional rudder deflection angle δ1 caused by the rotation of the rigid body and the additional rudder deflection angle δ2 caused by the elastic deformation, the total control rudder deflection angle is:

[0052]

[0053] In the formula, for The i-th order natural vibration mode at point;

[0054] Substituting equations (8)(9)(11)(12)(13)(14) into equations (4)(5), we can obtain the normal force and pitch moment respectively:

[0055]

[0056]

[0057] In the formula, y P is the Y-axis coordinate of the engine thrust action point with the leading edge of the aircraft as the coordinate origin. Then, the distribution of the time-varying normal force along the fuselage direction f(x,t) is:

[0058]

[0059] Substituting formula (27) into formula (7), we get:

[0060]

[0061] Substituting formula (28) into formula (3), we get:

[0062]

[0063] In the formula, i and j are the orders of the natural vibration modes, and q i Equivalent to q i (t), q j Equivalent to q j (t);

[0064]

[0065]

[0066] Furthermore, in step 2-4, let D 4ij =0,D 5ij =0, i≠j, and the Lagrangian treatment is performed on the center of mass translation motion equation and the center of mass rotation equation in formula (3), and the rigid-elastic coupling dynamic equation of the large dynamic pressure aircraft used for control system design is obtained:

[0067]

[0068] in:

[0069]

[0070] In the formula, is the derivative of the lift coefficient with respect to the angle of attack, is the derivative of the lift coefficient with respect to the pitch rudder deflection angle, a 22 ,a 23 ,a 25 ,a 34 ,a 35 is the rigid body dynamic coefficient, A 1i ,A 2i ,B 1i ,B 2i represents the influence of elastic vibration mode on the translation of center of mass and rotation around center of mass, D 1i ,D 2i ,D 3i ,D 4ij ,D 5ij is the generalized force coefficient of the i-th order natural vibration, which includes the coupling terms of the generalized coordinates and their derivatives. It can be found from the above coefficient form that there is mutual coupling between rigid body motion and elastic vibration. The coupling effect of rigid body mode on elastic mode is mainly reflected in the generalized force, and the coupling effect of elastic mode on rigid body mode is mainly reflected in the changes of thrust, aerodynamic force and aerodynamic moment.

[0071] Furthermore, in the third step, the sliding surface considering the rigid-elastic coupling is designed as:

[0072]

[0073] Where k 1,i ,k 2,i ,k 3,i , k4 is the design parameter, sgn() is the sign function, e is the pitch angle tracking error, is the pitch angle command;

[0074]

[0075] The sliding mode equivalent control law is designed as follows:

[0076]

[0077] The switching control law is designed to be the super-helical sliding mode form shown in formula (45):

[0078] δ z,sw =-k5s-k6sig(s)+∫(-k7sgn(s)-k8s) (45)

[0079] In the formula, k5, k6, k7, k8 are all design parameters, and sig(s) is expressed as:

[0080] sig(s)=||s|| 0.5 sgn(s) (46)

[0081] The controller design is:

[0082]

[0083] Furthermore, in the fourth step, a neural network observer is introduced to observe the coupled modal signal q of the aircraft i , that is, the generalized coordinates q of the i-th elastic mode i Realize online estimation of each mode;

[0084] The input layer of the neural network observer is the pitch angle of the aircraft, the number of nodes is 1, and the output layer is the modal signal q i , the number of nodes is n, the hidden layer is set to a 3-layer structure, namely P1, P2, P3 layers, the number of nodes is n p1 ,n p2 ,n p3 ; The activation function vectors of the three hidden layers of the neural network are H P1 (X), H P2 (X), H P3 (X); X is the neural network input layer vector. Since the number of input layer nodes is 1, X is the pitch angle of the aircraft.

[0085] The activation function vector H of the P1 layer of the neural network P1 (X), select Gaussian function as activation function, H P1 (X) Each element expression is:

[0086]

[0087] In the formula, h P1,m (X) is the vector H P1 The mth element in (X), C m is the center vector of the Gaussian function, b mis the base width parameter of the Gaussian function;

[0088] Data transfer between hidden layers of neural networks is achieved through weight matrices:

[0089]

[0090] Where W1 and W2 are the weight matrices between hidden layer P1 and hidden layer P2, and between hidden layer P2 and hidden layer P3 respectively;

[0091] According to the universal approximation theorem, the modal signal q i The neural network state observer is expressed as:

[0092] q i =W *T H P3 (X)+z1 (50)

[0093] Where W * is the ideal weight matrix of the output layer, z1 is the approximation error of the neural network;

[0094] The output layer weight matrix adaptive law update method is used to calculate the derivative of the actual output layer weight matrix

[0095]

[0096] In the formula for The adaptive update coefficient matrix is ​​of order, e is the pitch angle tracking error;

[0097] By introducing a robust control term into the neural network observer to reduce the influence of the neural network approximation error, the observed value of the modal signal is obtained. for:

[0098]

[0099] In the formula, k9 is the design parameter, z1 is the neural network approximation error, is the actual weight matrix of the output layer; the final super-helical sliding mode closed-loop control law based on the intelligent online estimation of rigid-elastic coupling mode is:

[0100]

[0101] By applying the above technical solution, the present invention has the following beneficial effects:

[0102] (1) The present invention provides an aerodynamic / thermal / structural / control multi-field coupling calculation and control method for a high dynamic pressure flight environment, the calculation and control method involves a numerical simulation calculation process of aerodynamic / thermal / structural multi-field coupling characteristics, and a dynamic modeling and control design method considering the influence of multi-field coupling characteristics on the aircraft; the calculation and control method realizes the coupling analysis of structural elastic deformation and aerodynamic / thermal loads through the "improved spatial discrete format CFD + CSD based on load distribution optimized grid" coupling numerical simulation calculation method; by introducing rigid-elastic coupling state variables in the aircraft dynamics modeling, an aircraft rigid-elastic coupling dynamic model that can accurately characterize the influence of multi-field coupling is established; by designing a sliding surface that can suppress the influence of multi-field coupling, and combining the rigid-elastic coupling mode intelligent online estimation and super-helical sliding mode idea, the control performance under the complex influence of rigid-elastic coupling is guaranteed. Therefore, the present invention has important reference significance for the analysis of aircraft multi-field coupling characteristics and control system design in a high dynamic pressure disturbance flight environment.

[0103] (2) The fluid domain model grid and the solid domain model grid of the present invention share geometric shape boundaries, which can adapt to the harsh conditions faced by high dynamic pressure aircraft, such as higher resistance, higher heat flow, greater load, and smaller aircraft design margin, and can accurately obtain the multi-field coupling characteristics under high dynamic pressure flight conditions.

[0104] (3) During the numerical simulation of the fluid domain model grid, the present invention improves the spatial discrete format of the numerical method of the simulation process. The improved spatial discrete format enhances the ability to capture complex flow structure characteristics, thereby significantly improving the accuracy of numerical simulation under high dynamic pressure flight conditions, and can more accurately obtain aerodynamic and aerodynamic thermal data under high dynamic pressure flight conditions.

[0105] (4) In the process of simulating and analyzing the multi-field coupling characteristics under high dynamic pressure flight conditions, the harsh areas of the aerodynamic and temperature fields use a direct simulation method to directly map the node distribution data of the aerodynamic and temperature fields on the grid boundaries of the solid domain model, and the non-harsh areas use a load migration method of conservative flux interpolation to interpolate the aerodynamic and temperature fields of the fluid domain model grid to the grid boundaries of the solid domain model; compared with the full-domain direct simulation method, this method can effectively improve the speed of multi-field coupling simulation analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0106] The included drawings are used to provide a further understanding of the embodiments of the present invention, which constitute a part of the specification, are used to illustrate the embodiments of the present invention, and together with the text description, explain the principles of the present invention. Obviously, the drawings in the following description are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0107] Figure 1 A flowchart of the aerodynamic / thermal / structural / control multi-field coupling calculation and control method for high dynamic pressure flight environment is given;

[0108] Figure 2 The structure diagram of super-helical sliding mode closed-loop control based on intelligent online estimation of rigid-elastic coupling modes is given. DETAILED DESCRIPTION

[0109] It should be noted that, in the absence of conflict, the embodiments in this application and the features in the embodiments can be combined with each other. The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present invention and its application or use. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0110] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0111] Unless otherwise specifically stated, the relative arrangement of the parts and steps described in these embodiments, numerical expressions and numerical values ​​do not limit the scope of the present invention. At the same time, it should be understood that, for ease of description, the sizes of the various parts shown in the accompanying drawings are not drawn according to the actual proportional relationship. The technology, method and equipment known to ordinary technicians in the relevant field may not be discussed in detail, but in appropriate cases, the technology, method and equipment should be regarded as a part of the authorization specification. In all examples shown and discussed here, any specific value should be interpreted as being merely exemplary, rather than as a limitation. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once a certain item is defined in an accompanying drawing, it does not need to be further discussed in subsequent drawings.

[0112] Embodiment 1:

[0113] This embodiment provides a large dynamic pressure aerodynamic / thermal / structural / control multi-field coupling calculation and control method, the specific steps are:

[0114] In the first step, the coupled numerical simulation method of "CFD (fluid mechanics calculation) in improved spatial discretization format + CSD (solid mechanics calculation) based on load distribution optimized grid" is used to calculate the elastic deformation data of the large dynamic pressure aircraft under the action of aerodynamic force and aerodynamic heat, and obtain the analysis results of the aerodynamic / aerodynamic heat / structure multi-field coupling characteristics;

[0115] The second step is to establish the aerodynamic model and dynamic model of the aircraft, carry out the aerodynamic model correction according to the multi-field coupling characteristics, introduce the elastic state quantity of the aircraft into the aerodynamic model polynomial, and introduce the elastic vibration equation into the dynamic model, and then establish the rigid-elastic coupling dynamic model of the large dynamic pressure aircraft;

[0116] The third step is to design a sliding surface that can ensure the stabilization of the rigid-elastic coupling mode, and obtain equivalent control items based on the rigid-elastic coupling dynamics model; further add a super-helical sliding mode switching control item to form a super-helical sliding mode closed-loop control law to ensure that the system quickly approaches the sliding surface and better overcomes chattering;

[0117] The fourth step is to design the output layer weight adaptation law of the BRF neural network state observer to realize the online estimation of the rigid-elastic coupling mode, and finally form a super-helical sliding mode closed-loop control law based on the intelligent online estimation of the rigid-elastic coupling mode for the high dynamic pressure flight environment.

[0118] Embodiment 2:

[0119] This embodiment is based on embodiment 1, see attached Figure 1 , the specific steps of the first step are as follows:

[0120] Step 1-1, for high dynamic pressure aircraft, fluid domain model grid and solid domain model grid are established at the same time, and the multi-field grids (i.e., fluid domain model grid and solid domain model grid) share geometric shape boundaries; in traditional aircraft aeroelastic analysis, the fluid domain model grid and the solid domain model grid usually use two sets of grids to carry out calculations separately, and data exchange between multi-field grids must be completed through interpolation, and the interpolation accuracy directly affects the accuracy of the final coupling result; high dynamic pressure aircraft face higher resistance, higher heat flow, and greater load, and the aircraft design margin is smaller. Therefore, in order to accurately obtain the multi-field coupling characteristics under high dynamic pressure flight conditions, the multi-field grids are processed with shared geometric shape boundaries.

[0121] Step 1-2, numerical simulation is performed on the fluid domain model grid to obtain high-precision numerical simulation results; in the process of numerical simulation of the fluid domain model grid, in order to more accurately obtain the aerodynamic force and aerodynamic heat data under high dynamic pressure flight conditions, the numerical method of the simulation process is improved in the spatial discrete format:

[0122] This embodiment adopts the improved spatial discretization format shown in formula (1), improves the traditional weighted form to enhance the ability to capture complex flow structure characteristics, thereby significantly improving the numerical simulation accuracy under high dynamic pressure flight conditions:

[0123]

[0124] in is the numerical flux, d k is the linear weight coefficient, ω k is the nonlinear weight coefficient, k=0,1,2.

[0125] Step 1-3, carry out load analysis based on the high-precision numerical simulation results obtained in step 1-2, and obtain aerodynamic and aerodynamic thermal data under high dynamic pressure flight conditions; in the process of load analysis, optimize the mesh of the fluid domain model and the mesh of the solid domain model, and the optimization process is: split the multi-field mesh area, retain the multi-field mesh of the force / heat load severe area (such as the leading edge of the aircraft, the air inlet, the wing surface, etc.), and interpolate and sparse the number of boundary mesh nodes in the non-severe area;

[0126] Steps 1-4, conduct aerodynamic thermal-structural heat transfer simulation analysis on the optimized multi-field grid to obtain the wall temperature field results of the solid domain model grid under aerodynamic heating conditions;

[0127] Step 1-5, based on the aerodynamic data obtained in step 1-3 and the temperature field results obtained in step 1-4, simulate and analyze the multi-field coupling characteristics under high dynamic pressure flight conditions to obtain the elastic deformation data of the aircraft; in the simulation and analysis process, the harsh area of ​​the aerodynamic and temperature fields uses a direct simulation method to directly map the node distribution data of the aerodynamic and temperature fields on the grid boundary of the solid domain model, and the non-harsh area uses a load migration method of conservative flux interpolation to interpolate the aerodynamic and temperature fields of the fluid domain model grid to the grid boundary of the solid domain model; compared with the full domain direct simulation method, the method in this step can effectively improve the speed of multi-field coupling simulation analysis;

[0128] Step 1-6, for the elastic deformation data of the aircraft, the topology-preserving dynamic mesh deformation technology is used to obtain the deformed fluid domain model mesh and solid domain model mesh, and elastic coupling calculation is iteratively performed until the final deformation converges, so as to obtain the multi-field coupling characteristics under the high dynamic pressure flight condition; wherein the multi-field coupling characteristics include: aerodynamic and aerodynamic thermal data before and after the elastic deformation of the aircraft, the stress and strain changes of the aircraft, etc.; this embodiment mainly uses the aerodynamic data before and after the elastic deformation of the aircraft;

[0129] Therefore, in the first step, the coupled numerical simulation calculation method of "CFD in the improved spatial discretization format involved in step 1-2 and CSD in the grid optimized according to load distribution involved in step 1-3" is used to calculate the elastic deformation data of the high dynamic pressure aircraft under the action of aerodynamic force and aerothermal force, and obtain the analysis results of the aerodynamic / aerothermal / structural multi-field coupling characteristics, specifically the aerodynamic data before and after the elastic deformation of the aircraft.

[0130] The specific steps of the second step are as follows:

[0131] Since the aircraft adopts a large slenderness ratio configuration, its elastic vibration is mainly manifested as the bending deformation of the body. In order to analyze the elastic vibration characteristics of the aircraft and quantify the influence of the elastic mode on the aerodynamic force, aerodynamic moment and the resulting elastic deformation of the aircraft, the aircraft is simplified to an Euler free beam model, and the moment equilibrium equation of the beam microelement can be obtained.

[0132] Step 2-1, using the modal superposition method, based on the elastic deformation data obtained in the first step, using the natural vibration mode Φ i The linear combination of (x), i=1,2,...,∞ expresses the elastic deformation function:

[0133]

[0134] Where: y(x,t) is the elastic deformation function of the body, q i (t) is the generalized coordinate of the i-th elastic mode;

[0135] Where i is the order of the natural vibration mode, x is the x-axis position of the point on the fuselage from the origin of the coordinate system with the leading edge of the aircraft as the origin, x is a continuous value, and Φ i (x) is the i-th order natural vibration mode at point x;

[0136] Based on the Lagrange equation, the separation of variables method is used to solve the elastic vibration equation, that is, the micro-element motion equation. The micro-element motion equation is combined with the center of mass translation equation and the rotation equation around the center of mass to obtain the rigid-elastic coupling dynamics equation group of the large dynamic pressure aircraft:

[0137]

[0138] Among them, the equation of motion of the microelement is the third equation in formula (3), the equation of motion of the center of mass translation is the first equation in formula (3), and the equation of motion of the rotation around the center of mass is the second equation in formula (3);

[0139] Where m is the mass of the aircraft, V is the flight speed, θ is the track angle, and F is the normal force. is the pitch angle, J z is the moment of inertia of the aircraft about the z-axis of the body coordinate system, Mz is the pitching moment, q i (t) is the generalized coordinate of the i-th elastic mode, ξ i and ω i are the structural damping ratio and natural frequency of the i-th elastic mode, Q i is the generalized force, M i is the generalized mass, F is the distributed aerodynamic force F y The concentrated force F generated by the thrust P The normal force, M z is the aerodynamic moment M zy With thrust torque M zP The pitching moment can be expressed as:

[0140] F=F y +F P (4)

[0141] M z =M zy +M zP (5)

[0142] The expressions of generalized mass and generalized force are:

[0143]

[0144] Where L is the total length of the aircraft fuselage, that is, the total length of the aircraft along the x-axis of the body coordinate system, m(x) is the distribution of the aircraft mass along the fuselage direction, and f(x, t) is the distribution of the time-varying normal resultant external force along the fuselage direction.

[0145] Step 2-2, based on the aerodynamic data obtained in the first step, it is assumed that the aerodynamic forces generated by the fuselage and wings of the aircraft are treated as distributed forces, while the control forces generated by the thrust and control surfaces are treated as concentrated forces;

[0146] The concentrated force F generated by the thrust P and thrust moment M zP They are:

[0147]

[0148] Where P is thrust, α is the rigid attack angle, and x p is the x-axis coordinate of the engine thrust point with the leading edge of the aircraft as the coordinate origin, l c is the distance from the engine nozzle section to the center of mass of the fuselage, l c =x p -x CM , x CM is the x-axis coordinate of the center of mass of the aircraft, Φ i (x p ) is x pThe i-th order natural vibration mode at point , and:

[0149]

[0150] Aerodynamic force F y for:

[0151]

[0152] F y The aerodynamic torque M generated zy for:

[0153]

[0154] Where ρ is the atmospheric density, S is the reference area of ​​the aircraft, and C y is the lift coefficient, α is the rigid angle of attack, α K is the local angle of attack; δ K To control the rudder deviation, δ z The rudder deflection required by the control signal, is the x-axis coordinate of the point of action of the control force generated by the rudder surface; Δ(·) is the Diracδ function.

[0155] Step 2-3, introduce the local angle of attack α K And control rudder deflection δ K , corrected aerodynamic force and aerodynamic moment: Among them, when correcting the aerodynamic force, it is realized in the form of the distribution of the time-varying normal force along the fuselage direction f(x,t);

[0156] The local angle of attack α K It consists of three parts: the rigid angle of attack α, the angle of attack α1 caused by the rotation of the rigid body, and the additional angle of attack α2 caused by elastic deformation. The total local angle of attack is:

[0157]

[0158] In the formula, is the pitch angle rate; Φ i (x l ) is the i-th order natural vibration mode at point x;

[0159] The control rudder deflection δ K It consists of three parts: the rudder deflection required by the control signal z , the additional rudder deflection angle δ1 caused by the rotation of the rigid body and the additional rudder deflection angle δ2 caused by the elastic deformation, the total control rudder deflection angle is:

[0160]

[0161] In the formula, for The i-th order natural vibration mode at point;

[0162] Substituting equations (8)(9)(11)(12)(13)(14) into equations (4)(5), we can obtain the normal force and pitch moment respectively:

[0163]

[0164]

[0165] In the formula, y P It is the Y-axis coordinate of the engine thrust action point with the leading edge of the aircraft as the coordinate origin.

[0166] Then, the distribution of the time-varying normal force along the fuselage direction f(x, t), that is, the normal distribution external force function f(x, t) applied to the aircraft is:

[0167]

[0168] Substituting formula (27) into formula (7), we get:

[0169]

[0170] Substituting formula (28) into formula (3), we get:

[0171]

[0172] Where i and j are the orders of the natural vibration modes, and throughout the text, q i Equivalent to q i (t), q j Equivalent to q j (t);

[0173]

[0174] Step 2-4, based on the small disturbance assumption, ignore the coupling between modes, that is, let D 4ij =0,D 5ij =0, i≠j, and the Lagrangian treatment is performed on the center of mass translation motion equation and the center of mass rotation equation in formula (3), and the rigid-elastic coupling dynamic equation of the large dynamic pressure aircraft used for control system design is obtained:

[0175]

[0176] in:

[0177]

[0178] In the formula, is the derivative of the lift coefficient with respect to the angle of attack, is the derivative of the lift coefficient with respect to the pitch rudder deflection angle, a22 ,a 23 ,a 25 ,a 34 ,a 35 is the rigid body dynamic coefficient, A 1i ,A 2i ,B 1i ,B 2i represents the influence of elastic vibration mode on the translation of center of mass and rotation around center of mass, D 1i ,D 2i ,D 3i ,D 4ij ,D 5ij is the generalized force coefficient of the i-th order natural vibration, which includes the coupling terms of the generalized coordinates and their derivatives. From the above coefficient form, it can be found that there is mutual coupling between rigid body motion and elastic vibration. The coupling effect of rigid body mode on elastic mode is mainly reflected in the generalized force, and the coupling effect of elastic mode on rigid body mode is mainly reflected in the changes of thrust, aerodynamic force and aerodynamic moment.

[0179] Therefore, in the second step, the aerodynamic model is corrected based on the multi-field coupling characteristics, and the elastic state of the aircraft is introduced into the aerodynamic model polynomial represented by formulas (15), (16), (28), that is, q j (t) and At the same time, the elastic vibration equation (i.e., the third equation of formula (3)) is introduced into the dynamic model, and then the rigid-elastic coupling dynamic model of the large dynamic pressure aircraft is established, i.e., formula (35).

[0180] The specific process of the third step is as follows:

[0181] On the basis of the traditional sliding surface, the elastic modal state quantity is introduced to ensure that the generalized coordinates of the elastic mode remain stable when the system is in the sliding surface, and to suppress the complex influence of multi-field coupling on the dynamic characteristics of the aircraft. The sliding surface design considering rigid-elastic coupling proposed in this embodiment is:

[0182]

[0183] Where k 1,i ,k 2,i ,k 3,i , k4 is the design parameter, sgn() is the sign function, e is the pitch angle tracking error, is the pitch angle command;

[0184]

[0185] The sliding mode equivalent control law is designed as follows:

[0186]

[0187] Combined with the idea of ​​super-helical sliding mode, in order to speed up the closed-loop system's approach to the sliding surface and better overcome chattering, the switching control law is designed as the following super-helical sliding mode form:

[0188] δ z,sw =-k5s-k6sig(s)+∫(-k7sgn(s)-k8s) (45)

[0189] In the formula, k5, k6, k7, k8 are all design parameters, and sig() is:

[0190] sig(s)=||s|| 0.5 sgn(s) (46)

[0191] The controller design is:

[0192]

[0193] Therefore, a sliding surface that can ensure the stabilization of the rigid-elastic coupling mode is designed, and the equivalent control term is obtained based on the rigid-elastic coupling dynamic model, that is, the equivalent control law expressed by formula (44). The super-helical sliding mode switching control term is further added, that is, the switching control law expressed by formula (45), to form a super-helical sliding mode closed-loop control law, that is, the controller expressed by formula (47).

[0194] The specific process of the fourth step is as follows:

[0195] Since the rigid-elastic coupled modes of large dynamic pressure aircraft are unmeasurable, it is difficult to achieve online estimation of each mode by system identification method. Therefore, it is considered to introduce a neural network observer to observe the coupled modal signal q of the aircraft. i , that is, the generalized coordinates q of the i-th elastic mode i The input layer of the neural network observer is the pitch angle of the aircraft, the number of nodes is 1, and the output layer is the modal signal q i , the number of nodes is n, the hidden layer is set to a 3-layer structure, namely P1, P2, P3 layers, the number of nodes is n p1 ,n p2 ,n p3 ; The activation function vectors of the three hidden layers of the neural network are H P1 (X), H P2 (X), H P3 (X); X is the neural network input layer vector. Since the number of input layer nodes is 1, X is the pitch angle of the aircraft.

[0196] The activation function vector H of the P1 layer of the neural network P1 (X), select Gaussian function as activation function, H P1 (X) Each element expression is:

[0197]

[0198] In the formula, h P1,m (X) is the vector H P1 The mth element in (X), C m is the center vector of the Gaussian function, b m is the base width parameter of the Gaussian function.

[0199] Data transfer between hidden layers of neural networks is achieved through weight matrices:

[0200]

[0201] Where W1 and W2 are the weight matrices between hidden layer P1 and hidden layer P2, and between hidden layer P2 and hidden layer P3, respectively.

[0202] According to the universal approximation theorem, the modal signal q i It can be expressed by the neural network state observer as:

[0203] q i =W *T H P3 (X)+z1 (50)

[0204] Where W * is the ideal weight matrix of the output layer, z1 is the approximation error of the neural network;

[0205] The output layer weight matrix adaptive law update method (such as the method shown in formula (51)) is used to calculate the derivative of the actual weight matrix of the output layer:

[0206]

[0207] In the formula n p3 The adaptive update coefficient matrix is ​​of order, e is the pitch angle tracking error;

[0208] By introducing a robust control term into the neural network observer to reduce the influence of the neural network approximation error, the observed value of the modal signal is obtained. for:

[0209]

[0210] In the formula, k9 is the design parameter, z1 is the neural network approximation error, is the actual weight matrix of the output layer;

[0211] The final super-helical sliding mode closed-loop control law based on the intelligent online estimation of rigid-elastic coupling mode is shown in formula (53), and the control structure is as follows: Figure 2 shown.

[0212]

[0213] Therefore, the output layer weight adaptive law of the BRF neural network state observer is designed, that is, the output layer weight matrix adaptive law update method shown in formula (51), and the calculation is obtained according to Calculate the observed value of the modal signal The online estimation of the rigid-elastic coupling mode is realized, and finally a super-helical sliding mode closed-loop control law based on the intelligent online estimation of the rigid-elastic coupling mode for the high dynamic pressure flight environment is formed, namely, formula (53).

[0214] For ease of description, spatially relative terms such as "above", "above", "on the upper surface of", "above", etc. may be used here to describe the spatial positional relationship between a device or feature and other devices or features as shown in the figure. It should be understood that spatially relative terms are intended to include different orientations of the device in use or operation in addition to the orientation described in the figure. For example, if the device in the accompanying drawings is inverted, the device described as "above other devices or structures" or "above other devices or structures" will be positioned as "below other devices or structures" or "below other devices or structures". Thus, the exemplary term "above" can include both "above" and "below". The device can also be positioned in other different ways (rotated 90 degrees or in other orientations), and the spatially relative descriptions used here are interpreted accordingly.

[0215] In addition, it should be noted that the use of terms such as "first" and "second" to limit components is only for the convenience of distinguishing the corresponding components. If not otherwise stated, the above terms have no special meaning and therefore cannot be understood as limiting the scope of protection of the present invention.

[0216] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A large dynamic pressure aerodynamic / thermal / structural / control multi-field coupling calculation and control method, characterized in that: The specific steps are: In the first step, the coupled numerical simulation method of "CFD in improved spatial discretization format + CSD with optimized grid based on load distribution" is used to calculate the elastic deformation data of the high dynamic pressure aircraft under the action of aerodynamic force and aerothermal force, and then the multi-field coupling characteristics of aerodynamic force / aerothermal force / structure are obtained; The second step is to establish the aerodynamic model and dynamic model of the aircraft, carry out the aerodynamic model correction according to the multi-field coupling characteristics, introduce the elastic state quantity of the aircraft into the aerodynamic model polynomial, and introduce the elastic vibration equation into the dynamic model, and then establish the rigid-elastic coupling dynamic model of the large dynamic pressure aircraft; The third step is to design a sliding surface that can ensure the stabilization of the rigid-elastic coupling mode, and obtain the equivalent control term based on the rigid-elastic coupling dynamics model; further add the super-helical sliding mode switching control term to form a super-helical sliding mode closed-loop control law; The fourth step is to design the output layer weight adaptation law of the BRF neural network state observer to realize the online estimation of the rigid-elastic coupling mode, and finally form a super-helical sliding mode closed-loop control law based on the intelligent online estimation of the rigid-elastic coupling mode for the high dynamic pressure flight environment.

2. A large dynamic pressure aerodynamic / thermal / structural / control multi-field coupling calculation and control method as claimed in claim 1, characterized in that: The specific steps of the first step are as follows: Step 1-1, for a large dynamic pressure aircraft, a fluid domain model grid and a solid domain model grid are simultaneously established, and the multi-field grids, that is, the fluid domain model grid and the solid domain model grid share geometric shape boundaries; Step 1-2, numerically simulating the fluid domain model grid to obtain numerical simulation results; in the process of numerically simulating the fluid domain model grid, improving the spatial discrete format of the numerical method of the simulation process; Step 1-3, carry out load analysis based on the numerical simulation results obtained in step 1-2 to obtain aerodynamic and aerodynamic thermal data under high dynamic pressure flight conditions; in the process of load analysis, optimize the mesh of the fluid domain model and the mesh of the solid domain model; Steps 1-4, conduct aerodynamic thermal-structural heat transfer simulation analysis on the optimized multi-field grid to obtain the wall temperature field results of the solid domain model grid under aerodynamic heating conditions; Step 1-5, based on the aerodynamic data obtained in step 1-3 and the temperature field results obtained in step 1-4, simulate and analyze the multi-field coupling characteristics under high dynamic pressure flight conditions to obtain the elastic deformation data of the aircraft; in the simulation and analysis process, the severe area of ​​the aerodynamic force and temperature field uses a direct simulation method to directly map the node distribution data of the aerodynamic force and temperature field on the grid boundary of the solid domain model, and the non-severe area uses a load migration method of conservative flux interpolation to interpolate the aerodynamic force and temperature field of the fluid domain model grid to the grid boundary of the solid domain model; Steps 1-6, for the elastic deformation data of the aircraft, use the topology-preserving dynamic mesh deformation technology to obtain the deformed fluid domain model mesh and solid domain model mesh, and iteratively perform elastic coupling calculations until the final deformation converges to obtain the multi-field coupling characteristics under high dynamic pressure flight conditions.

3. A large dynamic pressure aerodynamic / thermal / structural / control multi-field coupling calculation and control method as claimed in claim 2, characterized in that: In step 1-2, the improved spatial discretization format is shown in formula (1): in: is the numerical flux, d k is the linear weight coefficient, ω k is the nonlinear weight coefficient, k=0,1,2.

4. A large dynamic pressure aerodynamic / thermal / structural / control multi-field coupling calculation and control method as claimed in claim 2, characterized in that: In steps 1-3, the optimization process of the fluid domain model mesh and the solid domain model mesh is: split the multi-field mesh area, retain the multi-field mesh in the severe area of ​​force / thermal load, and interpolate and sparse the number of boundary mesh nodes in the non-severe area.

5. A large dynamic pressure aerodynamic / thermal / structural / control multi-field coupling calculation and control method as claimed in claim 1, characterized in that: The specific steps of the second step are as follows: Step 2-1, based on the elastic deformation data obtained in the first step, use the natural vibration mode Φ i The linear combination of (x), i=1,2,...,∞ expresses the elastic deformation function: Where: y(x,t) is the elastic deformation function of the body, q i (t) is the generalized coordinate of the i-th elastic mode; The rigid-elastic coupling dynamic equations of the high dynamic pressure aircraft are established; where i is the order of the natural vibration mode, x is the x-axis position of the point on the fuselage from the origin of the coordinate system in the body coordinate system, x is a continuous value, and Φ i (x) is the i-th order natural vibration mode at point x; Step 2-2, based on the aerodynamic data obtained in the first step, assuming that the aerodynamic forces generated by the fuselage and wings of the aircraft are treated as distributed forces, and the control forces generated by the thrust and control surfaces are treated as concentrated forces, the aerodynamic force F is calculated. y , aerodynamic moment M zy , the concentrated force F generated by the thrust P and thrust moment M zP ; Step 2-3, introduce the local angle of attack α K And control rudder deflection δ K , corrected aerodynamic force and aerodynamic moment; In step 2-4, based on the small disturbance assumption, the coupling between modes is ignored, and the rigid-elastic coupled dynamic equations of the large dynamic pressure aircraft established in step 2-1 are subjected to Lagrangian treatment to obtain the rigid-elastic coupled dynamic equations of the large dynamic pressure aircraft used for control system design.

6. A large dynamic pressure aerodynamic / thermal / structural / control multi-field coupling calculation and control method as claimed in claim 5, characterized in that: In step 2-1, based on the Lagrange equation, the separation of variables method is used to solve the micro-element motion equation. The micro-element motion equation is combined with the center of mass translation equation and the rotation equation around the center of mass to obtain the rigid-elastic coupled dynamic equation group of the high dynamic pressure aircraft: Where m is the mass of the aircraft, V is the flight speed, θ is the track angle, and F is the normal force. is the pitch angle, J z is the moment of inertia of the aircraft about the z-axis of the body coordinate system, M z is the pitching moment, q i (t) is the generalized coordinate of the i-th elastic mode, ξ i and ω i are the structural damping ratio and natural frequency of the i-th elastic mode, Q i is the generalized force, M i is the generalized mass, F is the distributed aerodynamic force F y The concentrated force F generated by the thrust P The normal force, M z is the aerodynamic moment M zy With thrust torque M zP The pitching moment can be expressed as: F=F y +F P (4) M z =M zy +M zP (5) The expressions of generalized mass and generalized force are: Where L is the total length of the aircraft fuselage, m(x) is the distribution of the aircraft mass along the fuselage direction, and f(x,t) is the distribution of the time-varying normal force along the fuselage direction.

7. A large dynamic pressure aerodynamic / thermal / structural / control multi-field coupling calculation and control method as claimed in claim 6, characterized in that: In step 2-2, the concentrated force F generated by the thrust P and thrust moment M zP They are: Where P is thrust, α is the rigid attack angle, and x p is the x-axis coordinate of the engine thrust point with the leading edge of the aircraft as the coordinate origin, l c is the distance from the engine nozzle section to the center of mass of the fuselage, l c =x p -x CM , x CM is the x-axis coordinate of the center of mass of the aircraft, Φ i (x p ) is x p The i-th order natural vibration mode at point , and: Aerodynamic force F y for: F y The aerodynamic torque M generated zy for: Where ρ is the atmospheric density, S is the reference area of ​​the aircraft, and C y is the lift coefficient, α is the rigid angle of attack, α K is the local angle of attack; δ K To control the rudder deviation, δ z The rudder deflection required by the control signal, is the x-axis coordinate of the point of action of the control force generated by the rudder surface; Δ(·) is the Diracδ function; In step 2-3, the local angle of attack α K It consists of three parts: the rigid angle of attack α, the angle of attack α1 caused by the rotation of the rigid body, and the additional angle of attack α2 caused by elastic deformation. The total local angle of attack is: In the formula, for Pitch rate, Φ i (x) is the i-th order natural vibration mode at point x; The control rudder deflection δ K It consists of three parts: the rudder deflection required by the control signal z , the additional rudder deflection angle δ1 caused by the rotation of the rigid body and the additional rudder deflection angle δ2 caused by the elastic deformation, the total control rudder deflection angle is: In the formula, for The i-th order natural vibration mode at point; Substituting equations (8)(9)(11)(12)(13)(14) into equations (4)(5), we can obtain the normal force and pitch moment respectively: In the formula, y P is the Y-axis coordinate of the engine thrust action point with the leading edge of the aircraft as the coordinate origin. Then, the distribution of the time-varying normal force along the fuselage direction f(x,t) is: Substituting formula (27) into formula (7), we get: Substituting formula (28) into formula (3), we get: In the formula, i and j are the orders of the natural vibration modes, and q i Equivalent to q i (t), q j Equivalent to q j (t); 8. A large dynamic pressure aerodynamic / thermal / structural / control multi-field coupling calculation and control method as claimed in claim 7, characterized in that: In step 2-4, let D 4ij =0,D 5ij =0, i≠j, and the Lagrangian treatment is performed on the center of mass translation motion equation and the center of mass rotation equation in formula (3), and the rigid-elastic coupling dynamic equation of the large dynamic pressure aircraft used for control system design is obtained: in: In the formula, is the derivative of the lift coefficient with respect to the angle of attack, is the derivative of the lift coefficient with respect to the pitch rudder deflection angle, a 22 ,a 23 ,a 25 ,a 34 ,a 35 is the rigid body dynamic coefficient, A 1i ,A 2i ,B 1i ,B 2i represents the influence of elastic vibration mode on the translation of center of mass and rotation around center of mass, D 1i ,D 2i ,D 3i ,D 4ij ,D 5ij is the generalized force coefficient of the i-th order natural vibration, which includes the coupling terms of the generalized coordinates and their derivatives. It can be found from the above coefficient form that there is mutual coupling between rigid body motion and elastic vibration. The coupling effect of rigid body mode on elastic mode is mainly reflected in the generalized force, and the coupling effect of elastic mode on rigid body mode is mainly reflected in the changes of thrust, aerodynamic force and aerodynamic moment.

9. A large dynamic pressure aerodynamic / thermal / structural / control multi-field coupling calculation and control method as claimed in claim 1, characterized in that: In the third step, the sliding surface considering the rigid-elastic coupling is designed as: Where k 1,i ,k 2,i ,k 3,i , k4 is the design parameter, sgn() is the sign function, e is the pitch angle tracking error, is the pitch angle command; The sliding mode equivalent control law is designed as follows: The switching control law is designed to be the super-helical sliding mode form shown in formula (45): δ z,sw =-k5s-k6sig(s)+∫(-k7sgn(s)-k8s) (45) In the formula, k5, k6, k7, k8 are all design parameters, and sig(s) is expressed as: sig(s)=||s||s 0.5 sgn(s) (46) The controller design is:

10. A large dynamic pressure aerodynamic / thermal / structural / control multi-field coupling calculation and control method as claimed in claim 9, characterized in that: In the fourth step, a neural network observer is introduced to observe the coupled modal signal q of the aircraft. i , that is, the generalized coordinates q of the i-th elastic mode i Realize online estimation of each mode; The input layer of the neural network observer is the pitch angle of the aircraft, the number of nodes is 1, and the output layer is the modal signal q i , the number of nodes is n, the hidden layer is set to a 3-layer structure, namely P1, P2, P3 layers, the number of nodes is n p1 ,n p2 ,n p3 ; The activation function vectors of the three hidden layers of the neural network are H P1 (X), H P2 (X), H P3 (X); X is the neural network input layer vector. Since the number of input layer nodes is 1, X is the pitch angle of the aircraft. The activation function vector H of the P1 layer of the neural network P1 (X), select Gaussian function as activation function, H P1 (X) Each element expression is: In the formula, h P1,m (X) is the vector H P1 The mth element in (X), C m is the center vector of the Gaussian function, b m is the base width parameter of the Gaussian function; Data transfer between hidden layers of neural networks is achieved through weight matrices: Where W1 and W2 are the weight matrices between hidden layer P1 and hidden layer P2, and between hidden layer P2 and hidden layer P3 respectively; According to the universal approximation theorem, the modal signal q i The neural network state observer is expressed as: q i =W *T H P3 (X)+z1 (50) Where W * is the ideal weight matrix of the output layer, z1 is the approximation error of the neural network; The output layer weight matrix adaptive law update method is used to calculate the derivative of the actual output layer weight matrix Where Γ1, Γ2,…, for The adaptive update coefficient matrix is ​​of order, e is the pitch angle tracking error; By introducing a robust control term into the neural network observer to reduce the influence of the neural network approximation error, the observed value of the modal signal is obtained. for: In the formula, k9 is the design parameter, z1 is the neural network approximation error, is the actual weight matrix of the output layer; The final super-helical sliding mode closed-loop control law based on the intelligent online estimation of rigid-elastic coupling modes is:

Images